{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:AGXSE5JWYX6K5OZRKBLN34PISS","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"f9ffbe979d80f8ad85588003f7357a6278fbbf583726809f997c33d9f185df8b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-11T16:38:11Z","title_canon_sha256":"a2c12a534db8513c14260db501d325e3cfe97d9d4db39c4e64f337af874b7071"},"schema_version":"1.0","source":{"id":"2310.07641","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.07641","created_at":"2026-07-05T08:08:18Z"},{"alias_kind":"arxiv_version","alias_value":"2310.07641v2","created_at":"2026-07-05T08:08:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.07641","created_at":"2026-07-05T08:08:18Z"},{"alias_kind":"pith_short_12","alias_value":"AGXSE5JWYX6K","created_at":"2026-07-05T08:08:18Z"},{"alias_kind":"pith_short_16","alias_value":"AGXSE5JWYX6K5OZR","created_at":"2026-07-05T08:08:18Z"},{"alias_kind":"pith_short_8","alias_value":"AGXSE5JW","created_at":"2026-07-05T08:08:18Z"}],"graph_snapshots":[{"event_id":"sha256:cca51f9eb7a99af6b2fdc406160dd4976aaa34e11e87ceaac90eeef646255776","target":"graph","created_at":"2026-07-05T08:08:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2310.07641/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As research in large language models (LLMs) continues to accelerate, LLM-based evaluation has emerged as a scalable and cost-effective alternative to human evaluations for comparing the ever increasing list of models. This paper investigates the efficacy of these ``LLM evaluators'', particularly in using them to assess instruction following, a metric that gauges how closely generated text adheres to the given instruction. We introduce a challenging meta-evaluation benchmark, LLMBar, designed to test the ability of an LLM evaluator in discerning instruction-following outputs. The authors manual","authors_text":"Danqi Chen, Jiatong Yu, Tanya Goyal, Tianyu Gao, Yu Meng, Zhiyuan Zeng","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-11T16:38:11Z","title":"Evaluating Large Language Models at Evaluating Instruction Following"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.07641","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:132a604d30d3f08a99332258b446d49a9df4f184ef35dcaff7df96c1b2c939b2","target":"record","created_at":"2026-07-05T08:08:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"f9ffbe979d80f8ad85588003f7357a6278fbbf583726809f997c33d9f185df8b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-11T16:38:11Z","title_canon_sha256":"a2c12a534db8513c14260db501d325e3cfe97d9d4db39c4e64f337af874b7071"},"schema_version":"1.0","source":{"id":"2310.07641","kind":"arxiv","version":2}},"canonical_sha256":"01af227536c5fcaebb315056ddf1e89493cb8619bfd522a35d52501c24df4372","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"01af227536c5fcaebb315056ddf1e89493cb8619bfd522a35d52501c24df4372","first_computed_at":"2026-07-05T08:08:18.428785Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:08:18.428785Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"J2p0TqGzMOT+BewKpmiB1yJ6/Ss4vkUhBI/ooKeY1DmzYstJL1gk9NEeoj8Gsl6VmVRC8gt4kkQG7VZU+mSOCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:08:18.429248Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.07641","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:132a604d30d3f08a99332258b446d49a9df4f184ef35dcaff7df96c1b2c939b2","sha256:cca51f9eb7a99af6b2fdc406160dd4976aaa34e11e87ceaac90eeef646255776"],"state_sha256":"967fdb999d65cf68eaba767fe1b99e29e22340ca18ccbee51f0bbe7876a52e1c"}